Metastasis is the leading cause of cancer death. Of all the processes involved in tumorigenesis, local invasion and the formation of metastases are the most clinically relevant, but the least understood. Intratumoral hypoxia, found in the majority of solid tumors, is associated with an increased risk of metastasis and treatment failure. Cancer cells adapt to the hypoxic microenvironment by increasing the activity of the hypoxia-inducible factors (HIF-1 and HIF-2). The mechanisms that drive HIF- regulated metastasis must be determined in order to identify effective treatment strategies with the potential to block metastasis. Our preliminary data showed that HIF expression promotes collagen deposition in vivo and in vitro in a HIF-1? dependent manner in both cancer and fibroblast cells by the transcriptional upregulation of collagen prolyl and lysyl hydroxylases. We also showed that collagen hydroxylase enzymes were essential for the spontaneous metastasis of breast cancer cells to the lung and lymph nodes of mice. On the other hand, we found that hypoxia induced the expression of integrin receptors in an ECM-independent manner. These observations led us to propose a model in which HIFs simultaneously induce the production of ECM components (ligands) and their integrin (receptors) to potentiate downstream signaling events which synergistically enhance metastasis. The proposed research will test this model by generating physiological ECM substrates which recapitulate the composition of ECM in vivo. In aim 1, we will test the hypothesis that HIF-1 or HIF-2 transcriptionally regulates the expression of several integrin subunits in cancer cells. During the mentored K99 phase, we will identify the HIF-dependent pattern of integrin gene expression under hypoxia. During the independent R00 phase, the mechanism of HIF-regulation of integrins will be determined. In aim 2, we will test the hypothesis that hypoxia-induced and HIF-dependent integrin expression causes an increase in ECM adhesion, motility, invasion and matrix contraction in a 3D culture model system using novel biophysical assays. During the mentored K99 phase, cell lines will be generated to counteract the effect of hypoxia on integrin expression and tested in the assays described. In aim 3, we will test the hypothesis that culturing breast cancer cells on ECM produced under hypoxic conditions will potentiate downstream integrin signaling. In the K99 phase, we will determine whether integrin expression is potentiated by interactions with a hypoxic ECM. In the R00 phase, we will test the hypothesis that some of the integrins induced under hypoxic conditions are required for HIF-induced metastasis. With this information in hand, during the R00 phase, we will systematically evaluate each step in the metastatic cascade using animal models to determine which steps require integrin(s) expression by breast cancer cells (aim 4). Taken together, we hope this data will lead to novel strategies for the treatment of metastatic breast cancer. The unique environment at Johns Hopkins has many advantages that will serve to support my training and research plans as well as my future scientific career. My primary mentor, Dr. Semenza and co-mentor, Dr. Denis Wirtz, are leaders in their respective fields. Their leadership together with an intensive career development training plan and the K99 award will facilitate my transition to a successful independent scientist.